Current approaches to transfer genes in vivo employ either recombinant viral vectors or non-viral delivery systems. Adeno-associated viral (AAV) vectors are nonpathogenic integrating vectors that infect both dividing and non-dividing cells. Since all viral genes are removed (96% of the viral genome), packaging of foreign DNA up to 5 kb can be incorporated into these vectors. Recently, we demonstrated transduction of rAAV vector expressing b-galactosidase after direct injection in rodent muscle. Vector DNA and transduced gene expression was detected for over one year without significant immune response. There was no evidence of vector toxicity in any animal treated suggesting the ability to safely and stably transfer sequences into muscle as an attractive platform for gene therapy. These studies have been extended into a large animal model (hemophilic FIX dog) using FIX sequences in place of b-galactosidase. Evidence for stable gene expression now out to 16 weeks suggest successful long term vector delivery. Analysis of vector spread was localized to site of injection and draining lymph node with infiltrating immune cells at the site of injection. Ironically, this infiltration correlated with residual inactivated Ad helper virus and not AAV transduced cells, suggesting further improvement in vector production is still required, a focus of this grant. Analysis of gene expression follows an atypical expression curve with onset of expression 10 to 21 days after vector delivery. In addition, gene expression appears to climb with time suggesting that molecular events that may be related to viral replication are taking place. These observation of stable Iong term gene expression, assumed to be obtained by vector integration, is in contrast to low level replication over four months. While these results demonstrate the first example of AAV productive gene therapy in a large animal, better elucidation of the molecular fate of the vector genome is required in order for safe translation into a clinical setting. Finally, these results strongly suggest that muscle is an attractive site for AAV transduction, and that further analysis of vector preps carrying various modified FIX genes devoid of helper Ad proteins will provide meaningful preclinical information that should facilitate AAV gene therapy for this genetic disorder.